Cross Layer Designs for OFDMA Wireless Systems with Heterogeneous Delay Requirements

This paper investigates a cross layer scheduling scheme for OFDMA wireless system with heterogeneous delay requirements. Unlike most existing cross layer designs which take a decoupling approach, our design considers both queueing theory and information theory in modeling the system dynamics. The cr...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE International Conference on Communications (2003) Jg. 12; S. 5325 - 5330
Hauptverfasser: Wing Hui, David Shui, Nang Lau, Vincent Kin, Lam, Wong Hing
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.06.2006
Schlagworte:
Base stations
Constraint optimization
Cross layer design
Delay systems
Design optimization
Information theory
Optimal scheduling
Power system modeling
Queueing analysis
Throughput
ISSN:1550-3607
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper investigates a cross layer scheduling scheme for OFDMA wireless system with heterogeneous delay requirements. Unlike most existing cross layer designs which take a decoupling approach, our design considers both queueing theory and information theory in modeling the system dynamics. The cross layer design is formulated as an optimization of total system throughput, subject to individual user's delay constraint and total base station transmit power constraint. The optimal scheduling algorithm for the delay-sensitive cross layer optimization is to dynamically allocate radio resources based on users' channel state information, source statistics and delay requirements. Specifically, optimal power allocation was found to be multilevel water-filling where urgent users have higher water-filling filinglevels, while optimal subcarrier allocation strategy is shown to be achievable by low complexity greedy algorithm. Simulation results also show the proposed jointly optimal power and subcarrier allocation policy can provide substantial throughput gain with all delay constraints being satisfied.
ISSN:1550-3607
DOI:10.1109/ICC.2006.255507